Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | saqlain raza <bhatti_sb@yahoo.com> |
To | "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu> |
Subject | Re: st: Exclusion restriction with mprobit |
Date | Thu, 3 Nov 2011 14:20:58 -0700 (PDT) |
I am using two binary dependent variables and a set of independent covariates. For my case, the errors are correlated with each other. What do you suggest? Saqlain RAZA France ________________________________ From: Maarten Buis <maartenlbuis@gmail.com> To: statalist@hsphsun2.harvard.edu Sent: Wednesday, November 2, 2011 9:25 PM Subject: Re: st: Exclusion restriction with mprobit On Wed, Nov 2, 2011 at 9:06 PM, saqlain raza wrote: > I am trying to use "mprobit" to estimate a multinomial probit model with four alternatives. > As far as I know, what we need for identifying MNP model is some set of proper "exclusion > restrictions," (as illustrated by Keane (1992)), not the suggested format of the data itself. That is only necesary when your multinomial probit model includes correlations between the different error terms. The -mprobit- command in Stata sets these correlations at 0, imposing the equivalent of the IAA assumption to a multinomial probit model. It is this assumption that identifies the model rather than an exclusion restriction. Hope this helps, Maarten -------------------------- Maarten L. Buis Institut fuer Soziologie Universitaet Tuebingen Wilhelmstrasse 36 72074 Tuebingen Germany http://www.maartenbuis.nl -------------------------- * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/ * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/